Related Experiment Videos
Estimation of respiration-induced noise fluctuations from undersampled multislice fMRI data
L R Frank1, R B Buxton, E C Wong
1Department of Radiology, University of California at San Diego, San Diego, California, USA. lfrank@ucsd.edu
Magnetic Resonance in Medicine
|April 3, 2001
Summary
Functional MRI data noise from breathing can be reduced using existing multislice imaging. This method leverages unaliased noise information to improve functional activation accuracy and significance determination.
Area of Science:
- Neuroimaging
- Biomedical Engineering
- Signal Processing
Background:
- Functional MRI (fMRI) data are susceptible to structured noise from physiological processes.
- High-frequency physiological noise, particularly respiratory fluctuations, can alias into the fMRI signal.
- This aliasing compromises the accurate estimation and statistical significance of brain activity.
Purpose of the Study:
- To demonstrate that unaliased noise information exists within standard multislice fMRI data.
- To show this information can be utilized for noise reduction.
- To specifically address noise from high-frequency respiratory fluctuations.
Main Methods:
- Analysis of multislice fMRI time series data.
- Identification and utilization of unaliased noise components.
- Development of a method to estimate and reduce respiratory-related noise.
Main Results:
- Demonstration of available unaliased noise information in multislice fMRI data.
- Successful estimation of noise due to high-frequency respiratory fluctuations.
- Validation of a method for noise reduction using this information.
Conclusions:
- Unaliased noise information in multislice fMRI data can be effectively leveraged.
- This approach allows for the estimation and reduction of aliased physiological noise.
- Improved fMRI data quality enhances the reliability of functional activation analysis.